GA-GRGAT: A novel deep learning model for high-speed train axle temperature long term forecasting

Man, Jie; Dong, Hui; Gao, Jiayang; Zhang, Jun; Jia, Limin; Qin, Yong

doi:10.1016/j.eswa.2022.117033

Cited by 6 publications

(1 citation statement)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The accuracy of this method was found to be higher than that of using the BP neural network alone. Man et al [22] combined GAT and GAN into a method for predicting long-term axial temperature. The GRGAT framework was used as a spatio-temporal fusion for temperature prediction, and the GAN network was used to construct a time-conditional sequence with the GRGAT framework after analyzing the cyclical changes in axial temperature, and the historical axial temperature information was fused to improve the long-term prediction accuracy of the GA-GRGAT model.…”

Section: Introductionmentioning

confidence: 99%

Test Method for Rapid Prediction of Steady-State Temperature of Outer Rings of Bearings under Grease Lubrication Conditions

Xia,

Yang,

et al. 2024

Lubricants

View full text Add to dashboard Cite

Temperature has a great influence on the stability of bearing performance. For the study of bearing steady-state temperature, this paper proposes a test method to quickly predict the steady-state temperature of the outer ring of a bearing, which solves the problems in traditional theoretical calculations and simulation analysis methods such as the large number of calculations, complex models, and large errors. Firstly, a mathematical prediction model is established according to the bearing temperature-rise law; then, a bearing steady-state temperature detection device is designed; and finally, the prediction model parameters are solved according to the experimental results, and experimental verification is carried out. It is shown that the prediction model has high accuracy under different load and speed conditions, and the error between the predicted steady-state temperature and the tested steady-state temperature is less than 0.7 °C. This prediction method reduces the single test time of the same speed to 60 min, which greatly improves the efficiency of the temperature detection test. The steady-state temperature model has important theoretical significance in guiding the study of the limiting speed of bearings.

show abstract

Section: Introductionmentioning

confidence: 99%